Insight into β-hairpin stability: A structural and thermodynamic study of diastereomeric β-hairpin mimeticsy

Article abstract of DOI:10.1039/b111241d

Two diastereomers of a model β-hairpin peptide mimetic were synthesized and studied with a combination of experimental (NMR, X-ray, CD, MS, IR) and computational methods (Monte Carlo/molecular mechanics calculations). The secondary structure-stabilizing effects of hydrophobic interactions and hydrogen bonding were investigated. Comparison of the extent of folded hairpin population in non-competitive, polar aprotic, and polar protic solvents illustrates the critical role of intramolecular hydrogen bonding on hairpin stability. Investigation of ¹H NMR melting curves of the diastereomeric compounds in a variety of solvents allowed an evaluation of the role of hydrophobic effects on secondary structure stabilization to be made.

Full text of DOI:10.1039/b111241d

View Article Online / Journal Homepage / Table of Contents for this issue
Insight into b-hairpin stability: a structural and thermodynamic study
of diastereomeric b-hairpin mimeticsy
a,b
Mate Erdelyi, Vratislav Langer, Anders Karlen and Adolf Gogoll*
c
b
a
´ ´
´
´
a
Department of Organic Chemistry, Uppsala University, Box 531, 751 21 Uppsala, Sweden.
E-mail: adolf@kemi.uu.se; Fax: +46 18 512 524
Department of Medicinal Chemistry, Uppsala University, Box 574, 751 23 Uppsala, Sweden.
Fax: +46 18 471 4474
Department of Environmental Inorganic Chemistry, Chalmers University of Technology, 412 96
Go¨teborg, Sweden. Fax: +46 31 772 2853
b
c
Received (in London, UK ) 10th December 2001, Accepted 23rd April 2002
First published as an Advance Article on the web 6th June 2002
Two diastereomers of a model b-hairpin peptide mimetic were synthesized and studied with a combination of
experimental (NMR, X-ray, CD, MS, IR) and computational methods (Monte Carlo/molecular mechanics
calculations). The secondary structure-stabilizing effects of hydrophobic interactions and hydrogen bonding
were investigated. Comparison of the extent of folded hairpin population in non-competitive, polar aprotic,
and polar protic solvents illustrates the critical role of intramolecular hydrogen bonding on hairpin stability.
1
Investigation of H NMR melting curves of the diastereomeric compounds in a variety of solvents allowed an
evaluation of the role of hydrophobic effects on secondary structure stabilization to be made.
Introduction
b-Hairpins have been shown to play a key role in vital as well
as in pathological processes. They frequently participate in
protein–protein,1 protein–RNA,2 and protein–DNA3 recogni-
tion. Prion diseases,4 Alzheimer’s disease,5 the IgE-mediated
allergic response,6 the interaction of bacterial cell surface-asso-
ciated protein with IgG,7 and HIV gp120 binding to human T-
cell surface protein CD48 are some examples involving the b-
hairpin secondary structure element in biological processes.
In spite of its significance, the principles underlying hairpin
stability are still unresolved, perhaps because of the general
difficulty of investigating hairpins, due to their low solubility
and high propensity to aggregate.9,10 Despite these obstacles,
Fig. 1 The investigated model b-hairpin mimetics.
several examples of naturally occuring b-hairpins11 and hair-
pin mimetics12 have been reported recently.
While it is known that several factors affect hairpin stabi-
lity,z their relative importance is not well understood. A gen-
eral difficulty when trying to evaluate the impact of one
particular stabilizing factor is the presence of several, possibly
cooperative, weak interactions.23c,26 A deeper understanding
of the influence of different types of non-covalent interactions
on conformational equilibria is desirable for both theoretical
and practical reasons. For a quantitative evaluation of such
interactions, we have investigated two diastereomers of a
model hairpin mimetic (Fig. 1). The similarity of these two dia-
stereomers made it possible to separately study the role of
hydrophobic interactions, since the effects of the turn region,
solvent, hydrogen bonding, electrostatic effects, and chain
length are identical. Such a selective examination of one
non-covalent, weak interaction in the presence of other coop-
erative factors has not been attempted previously and is, there-
fore, of general interest.
Gellman and co-workers have shown that the configuration
of the amino acids in the turn region affects hairpin stabili-
ty.16c,d However, many of the turn mimics reported are achiral
and, therefore, have no preferred twist.4b,9a,12,16e–i,23a,b,27b To
avoid complications possibly arising from the turn region,
the known rigid, achiral tolan turn mimic16e,f was chosen for
this investigation, focusing on the effect of non-covalent side-
chain interactions on hairpin conformational stability. For
our model study, we selected the shortest possible amino acid
strands which allow the formation of two intramolecular
hydrogen bonds. By selecting small, nonpolar, nonaromatic
amino acid side chains, we could simultaneously minimize
the size of the model system as well as the interactions between
the turn mimetic and the peptide strands. To date, this is the
y Electronic supplementary information (ESI) available: temperature
and concentration-dependent chemical shifts and melting curves of
the investigated molecules in different solvents and details of the X-
ray analysis. See http://www.rsc.org/suppdata/nj/b1/b111241d/
z For example, the b-sheet-forming tendencies of amino acids,8b,13,14
the length of the peptide strands,15,16 the type of b-turn8b,9c,11b,13c,17
or the characteristics of the b-turn mimetic,12b,16c,d the side-chain prop-
erties,8b,18 context effects,8b,13a,19 steric factors,20 the role of interchain
hydrogen bonds,16c,21,22 b-branching in amino acid side chains, hydro-
phobic cluster formation,8b,13,23,24 and electrostatic interactions.9c,25
834
New J. Chem., 2002, 26, 834–843
DOI: 10.1039/b111241d
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2002

View Article Online
=
Scheme 1 Outline of the synthesis of the diastereomeric hairpin mimetics. Reagents and conditions: (a) Me SiC CH, Pd(PPh ) Cl , CuI, Et NH,
=
3
3 2
2
2
120 ꢂC, 5 min, 98%; (b) KFꢀ2H2O, MeOH, r.t., 7 h, 97%; (c) methyl-2-iodobenzoate, Pd(PPh3)2Cl2 , CuI, Et2NH, 120 ꢂC, 5 min, 78%; (d) BSA,
i
i
HATU, Pr2NEt, (S)-2-acetylamino-3-methylbutyric acid, r.t., 72 h, 52%; (e) KOtBu, Et2O, 10 h, 99%; (f) HATU, Pr2NEt, (S)-2-amino-N-
methyl-propionamide, r.t., 1 h, 51%; (g) HPLC: 55% 7a, 45% 7b.
Fig. 2 Hydrogen bonds connecting molecules of 7a in an infinite rib-
bon in the a direction. Hydrogen bonds, intramolecular: H(11a)–
Fig. 3 Hydrogen bonds connecting conformers A and B of 7a in two
separate ribbons parallel to the a direction. Hydrogen bonds in A, see
˚
O(27a) ¼ 2.33, H(33a)–O(3a) ¼ 2.03 A; intermolecular: H(4a)–
˚
O(32a) ¼ 2.07, H(28a)–O(10a) ¼ 2.16 A.
Fig. 2. Hydrogen bonds in B, intramolecular: H(11b)–O(27b) ¼ 2.43,
˚
H(33b)–O(3b) ¼ 1.95 A; intermolecular: H(4b)–O(32b) ¼ 2.01,
˚
H(28b)–O(10b) ¼ 2.17 A. End methyl groups and Val side-chain atoms
were omitted for clarity. Symbols refer to those in Table S7 (ESI).
first investigation of hairpins which differ in the configuration
of one of the peptide strands.
Conformational analysis by molecular mechanics calculations
Monte Carlo simulation combined with molecular mechanics
energy minimization in the OPLS-AA force field was used to
predict the conformation population for the two diastereomers
Results and discussion
in chloroform solution. Conformations within 25 kJmolꢁ1 of
Synthesis
the lowest energy minimum were examined. The obtained
Scheme 1 outlines the methodology used to prepare the (S)-
Val, (S)-Ala (7a) and (R)-Val, (S)-Ala (7b) derivatives of 2-
amido-20-carboxamidotolane (7).
structures were then classified as ‘‘hairpin’’ or ‘‘non-hairpin’’,
depending on the presence of two interchain hydrogen bonds.
The conformations identified as b-hairpin amounted to a popu-
lation of 15.6 and 20.7% at 298 K for diastereomers 7a and 7b,
respectively, using the Boltzman equation for the calculation.
X-Ray analysis
Conformational analysis in solution
The crystallographic analysis of diastereomer 7a (Fig. 2 and 3)
shows that the molecule has a high tendency to form 10 and
14-membered ring C=Oꢀ ꢀ ꢀH–N hydrogen bonds, as are com-
monly observed in b-hairpins. The asymmetric unit of the tri-
clinic crystal system contains two independent molecules,
showing a slight difference in hydrogen bond lengths. These
two conformers of 7a give rise to two distinct ribbon structures
(see ESI). No crystals suitable for X-ray analysis could be
obtained for 7b.
Where solubility in water was an issue, similarly to other b-
hairpin studies,9b,16,18,22 the investigations were made in
methanol, dimethylsulfoxide, and chloroform solutions.
Circular dichroism
The CD spectrum of diastereomer 7a recorded in methanol
shows the characteristic perturbation feature for tolan
New J. Chem., 2002, 26, 834–843
835

View Article Online
Fig. 4 The CD spectra of the hairpin diastereomers (methanol solu-
tion).
derivatives,16e,f as well as a positive band at 209 nm and an
intense negative band at 228 nm, typical of a b-turn28
(Fig. 4). We presume that the small size of the mimetic is
responsible for its CD spectrum resembling more that of
b-turns than that of b-hairpins.x In general, the CD spectra
of b-structures tend to depend on the length of the side chains
and the twist of the turn region.30 The reversed chirality at one
centre in diastereomer 7b gave rise to a spectrum in methanol
that resembled, apart from the intensities, the mirror image of
the spectrum of 7a. This feature is unusual for diastereomer
pairs and it suggests that the contributions of the chromo-
phores near to the valine alpha carbon are dominant over
the contribution of the chromophores near to the alanine
alpha carbon. The observed CD spectra allow the configura-
tional assignment of the two diastereomers, with the diastereo-
mer 7a showing a spectrum characteristic for b-structures
consisting of natural amino acids. This assignment of the abso-
lute configuration at Ca of valine was confirmed by X-ray and
1D NOE NMR experiments. Finally, the different magnitudes
of the ellipticities at 228 nm might suggest that the diaster-
eomer 7a has a slightly higher population of b-structure con-
formation than 7b.
Fig. 5 ESI mass spectra of (a) 7a and (b) 7b.
from FD and FAB MS investigations of diastereomeric tripep-
tide ester derivatives.32
IR spectra
The presence of intramolecular amide–amide hydrogen bond-
ing as an indicator of hairpin folding was probed by analysing
the amide N–H stretch region in chloroform solution. In this
spectral region, both the 7a and 7b diastereomers exhibited
an intense IR band at 3307–3308 cmꢁ1 (hydrogen-bonded
N–H vibration) and a less intense band at 3413 cmꢁ1 (non-
hydrogen-bonded N–H vibration).16d,27,29 These results indi-
cate that both hydrogen-bonded and non-bonded amide pro-
tons were present in the solutions of both diastereomers,
with no signs of any differences.
NMR investigation
1
Structural assignment. The H and 13C NMR spectra of 7a
and 7b have been fully assigned in several solvents using data
from COSY,33 HSQC,34 COLOC,35 NOESY,36 ROESY,37 and
gHMBC38 experiments. The 1H chemical shift assignments are
presented in Table 1. The absolute configuration of the two
hairpin diastereomers 7a and 7b was confirmed using 1D
NOE difference39 spectra (Fig. 6). These spectra were run on
CDCl3 solutions at room temperature after exchange of all
amide NH protons with deuterium. As expected, interstrand
NOEs were observed between the methyl groups of valine
and alanine for the 7a diastereomer, but not the 7b diastereo-
mer (Fig. 1). Similar effects were obtained by irradiation of
either one of the valine and alanine methyl groups. Interest-
ingly, these small effects could not be observed without deuter-
ium exchange. This further confirms the configurational
assignment made by CD spectroscopy and X-ray crystallo-
graphy.
Mass spectra
The two diastereomers show an unexpected difference in their
ESI mass spectra (Fig. 5). Whereas the spectrum of 7a contains
only the [M + H]+ and [M + Na]+ ions, the spectrum of 7b
shows, in addition to [M + 1]+ and [M + Na]+, several other
fragment ions, of which m/z ¼ 322 is the base peak. Such dif-
ferent behavior of diastereomers in mass spectra has been
reported before only in a very few cases.31 The differences in
the population equilibria of the possible conformations—
showing various extents of accessibility for ionization—is pre-
sumed to be responsible for the differences in the spectra of the
diastereomers. This assumption is supported by the results
Absence of aggregation behavior. Concentration dependency
of amide chemical shifts. The absence of intermolecular interac-
tions in solution was probed by investigating the concentration
dependency of DdNH for all amide protons. The variation of
DdNH as a function of concentration in chloroform, methanol
and DMSO solution was negligible (see ESI).
x Usually, the hairpin conformation in peptides containing natural L-
amino acids is indicated by a strong negative CD band at ca. 216–222
nm.16a,29
836
New J. Chem., 2002, 26, 834–843

View Article Online
Table 1 1H chemical shift assignments (d) for the side chains of b-hairpin mimetics 7a and 7b at 298 K in DMSO-d6 , CH3OH–CD3OD (2:1), and
CDCl3 solution
7a
7b
Residue
NHPh
DMSO-d6
CH3OH–CD3OD
CDCl3
DMSO-d6
CH3OH–CD3OD
CDCl3
9.47
7.91
8.77
8.21
4.58
1.30
2.51
4.88
2.11
0.90
0.88
1.92
9.54
8.02
8.54
8.21
4.68
1.41
2.54
5.04
2.20
0.96
0.95
1.99
9.34
7.89
7.08
6.54
5.32
1.48
2.76
5.50
2.33
1.05
0.95
2.08
9.38
7.77
8.72
8.17
4.63
1.29
2.57
4.82
2.12
0.89
0.88
1.89
9.54
8.02
8.54
8.21
4.68
1.41
2.55
5.04
2.20
0.96
0.95
1.99
9.68
7.87
6.96
6.64
5.26
1.36
2.81
5.64
2.25
0.93
0.87
2.11
NHCH
3
NHAla
NHVal
Ala-CHa
Ala-CH3
NHAla
CH3
Val-CHa
Val-CHb
Val-CHg1
Val-CHg2
COVal
CH3
Fig. 6 Aliphatic region of the 1D NOE difference spectra for the dia-
stereomers 7a and 7b, (400 MHz, 25 ꢂC, CDCl3 solution). (a) 7a, refer-
ence spectrum; (b) 7a, NOE difference spectrum, one Val-Me
irradiated; (c) 7b, reference spectrum; (d) 7b, NOE difference spectrum,
one Val-Me irradiated. Integrated steady state NOEs are shown below
the signals (in %, normalized to the irradiated methyl
group ¼ ꢁ300%).
Evidence for hairpin conformation in solution interstrand
NOEs. In CDCl3 , b-hairpin formation is indicated by the
observation of interstrand NOEs. The strongest NOEs were
observed between the alpha protons of valine and alanine
(Fig. 7). The NOEs between the signals of these protons were
much weaker in DMSO-d6 solution for both diastereomers,
indicating that the b-hairpin conformations are more popu-
lated in the non-competitive solvent CDCl3 . These NOEs
could not be studied in methanol solution due to signal
overlap.
Fig. 7 Interstrand NOEs observed in CDCl3 solution (400 MHz,
25 ꢂC).
the NHPh and the NHCH proton resonances of Ddsolv ¼ ꢁ0.3
3
ꢁ0.3 and ꢁ0.1 ppm, respectively, while the NHAla and NHVal
protons showed Ddsolv ¼ 1.8 and 1.6 ppm, respectively. When
changing the solvent from CDCl3 to methanol, the shift
changes for NHPh and NHCH were Ddsolv ¼ ꢁ0.3 and 0.1
3
Solvent effects on amide chemical shifts. In agreement with
the low energy conformation obtained by computation and
ppm, while for both the NHAla and the NHVal protons,
Ddsolv ¼ 1.6 ppm. This observation supports a b-hairpin con-
X-ray studies, the NHPh and NHCH protons (see Fig. 1 for
3
formation, in which the NHPh and NHCH protons are hydro-
3
assignment) for both diastereomers were much less affected
by changes in the solvent than were the signals due to the
NHAla and NHVal protons (Table 1). For diastereomer 7a,
upon changing the solvent from CDCl3 to DMSO-d6 , the che-
gen bonded, and NHAla and NHVal are not (Fig. 1).28
Amide proton temperature coefficients (Dd/DT). A compari-
son of the temperature coefficients for the amide protons
(Table 2) provided further evidence for folded b-hairpin con-
mical shifts of the NHPh and the NHCH signals changed by
3
Ddsolv ¼ 0.1 and 0.01 ppm, respectively, whereas the chemical
shifts of NHAla and NHVal changed by Ddsolv ¼ 1.6 ppm
formations in which protons NHPh and NHCH are intramole-
3
(Ddsolv ¼ dDMSO ꢁ dCDCl ). The same trends were observed
cularly hydrogen bonded.{
3
when changing the solvent from CDCl3 to methanol
[Ddsolv ¼ 0.2 (NHPh), ꢁ0.1 (NHCH ), 1.4 (NHAla) and 1.6
3
{ Temperature coefficients, (dT ꢁ dT )/(Thigh ꢁ Tlow), are obtained
ppm (NHVal)]. For the diastereomer 7b, the amide protons
showed similar behavior. Thus, changing the solvent from
CDCl3 to DMSO-d6 resulted in chemical shift changes for
high
low
as negative numbers, but are reported as positive values, in accordance
with the accepted literature procedure: H. Kessler, Angew. Chem.,
1982, 94, 509.
New J. Chem., 2002, 26, 834–843
837

View Article Online
Table 2 Amide temperature coefficientsa in DMSO-d6 , CH3OH–CD3OD (2:1), and CDCl3 solution
7a
7b
Residue
NHPh
DMSO-d6
CH3OH–CD3OD
CDCl3
DMSO-d6
CH3OH–CD3OD
CDCl3
3.1
4.5
5.9
5.3
2.6
4.8
7.3
7.7
0.4
4.2
1.4
1.7
2.6
3.6
5.5
5.1
2.5
4.8
7.1
7.6
0.7
4.1
1.7
1.8
NHCH
3
NHAla
NHVal
a
The amide proton temperature coefficients, Dd(NH)/DT (in ppb Kꢁ1) were measured for 8 mmol dmꢁ3 samples in DMSO over the temperature
range 298–418 K, in CH3OH–CD3OD (2:1) over the range 178–323 K, and in CDCl3 over the temperature range 298–333 K.
Table 3 Thermodynamic parameters for unfolding of b-hairpin mimetic 7a and 7b in DMSO-d6 and CH3OH–CD3OD (2:1) solutions
Solvent
Tm/K
DHm/kJ molꢁ1
DSm/J Kꢁ1 molꢁ1
DGꢂ298 K/kJ molꢁ1
7b
7a
7b
DMSO-d6
CH3OH–CD3OD
CH3OH–CD3OD
388 ꢃ 5.8
16.5 ꢃ 0.6
11.2 ꢃ 0.6
11.1 ꢃ 0.4
42.5 ꢃ 0.9
36.4 ꢃ 1.3
36.9 ꢃ 1.9
3.8 ꢃ 0.3
0.32 ꢃ 0.3
0.12 ꢃ 0.1
306.9 ꢃ 7.5
301.4 ꢃ 3.8
1
Results for DMSO-d6 solution. The H chemical shift mea-
In the polar, aprotic dimethylsulfoxide-d6 , the amide pro-
tons NHAla and NHVal have high temperature coefficients
( > 5 ppb Kꢁ1) in both diastereomers, showing that they are
solvent exposed. The amide proton NHPh has the lowest tem-
surements were made over the temperature range 298 to 418
K. Data for Ala-CH3 , Val-CH3 , Val-CHb , Val-CHa , and
NHAla
CH3
were investigated.** The analysis of these melting
perature coefficient. The NHCH protons in both diastereomers
curves lead to the conclusion that diastereomer 7b has its melt-
ing point (Tm) at approximately 388 K (Table 3), while the
melting point of diastereomer 7a is likely to be higher than
418 K, the highest temperature accessible for our NMR
equipment.yy By comparing the melting curves, we conclude
that the b-hairpin conformation of mimetic 7a is significantly
more stable in the polar aprotic solvent DMSO than is that
of its diastereomer 7b (DTm > 30 K).
3
show an intermediate behavior, indicating that they are in
equilibrium between hydrogen-bonded and non-hydrogen-
bonded states.28,29
In polar, protic methanol, the temperature coefficients of the
two diastereomers are very similar. The NHAla and NHVal
amide protons may be solvent accessible, while proton NHCH
3
is in equilibrium between hydrogen-bonded and non-hydro-
gen-bonded states. NHPh has a temperature coefficient of less
than than 3 ppb Kꢁ1 and could, therefore, be hydrogen
bonded.
Results for methanol solution. The chemical shifts of Val-
CH3 , CH3COVal, and CH3
NHAla
were measured over the tem-
In apolar, aprotic CDCl3 , the amide temperature coeffi-
cients are more difficult to interpret,16a,27a and can only be
understood in combination with other parameters, such as che-
mical shift (Table 1) and its solvent-dependent behavior
(Ddsolv , see preceding section). According to the classification
introduced by Nowick and Soth.,16a the protons NHAla and
NHVal are not hydrogen bonded in any of the diastereomers
(low d, low Dd/DT, high Ddsolv). Proton NHPh may be strongly
hydrogen bonded in both mimetic 7a and 7b (high d, low Dd/
DT, low Ddsolv). However, the NMR parameters of this proton
should be interpreted carefully because the phenyl ring
attached to NHPh affects its solvent accessibility and chemical
perature range 178 to 328 K.zz The calculated thermodynamic
parameters are reported in Table 3. They indicate that in the
polar protic solvent methanol, the two diastereomers have
similar stabilities within experimental error.
Results for CDCl3 solution. The melting curve was measured
over the temperature range 213 to 328 K. Analysis of the reci-
procal temperature curve suggests that the inflexion point of
the melting curve might be above the boiling point of CDCl3
and, therefore, the thermodynamic constants can not be calcu-
lated. This observation leads to the assumption that in the apo-
lar, aprotic solvent CDCl3 , both diastereomers are highly
folded.
shift. Proton NHCH could be in equilibrium between hydro-
3
gen-bonded and non-hydrogen-bonded states (intermediate d,
high Dd/DT, low Ddsolv). In CDCl3 solution, below room tem-
perature, both hairpin mimetics showed a tendency for self-
association, indicated by the rapid change in the temperature
coefficients of the NHAla and NHVal protons.k
Entropic and enthalpic contributions to b-hairpin stability
Comparison of the thermodynamic parameters of one of the
diastereomer hairpins in different solvents provides quantita-
tive information on the importance of interstrand hydrogen
bonding, and allows a comparison of the conformer-stabilizing
Thermodynamic analyses
** Analyses of Ala-CHa and CH3COVal were not performed because of
the small difference in chemical shift for these protons between the
folded and unfolded states (|dF ꢁ dU| < 0.03 ppm).40a,b
A common method for quantitative thermodynamic descrip-
tion of hairpin stability is investigation of the protein melting
curve.13b,25,26,40 We have therefore investigated the tempera-
ture dependency of chemical shifts of side-chain and alpha
protons.40a,b
yy The fact that the inflexion point of the melting curve of 7a could not
be observed ruled out the possibility of calculating exact thermody-
namic parameters for this diastereomer.
zz The NMR signals of alpha protons were omitted from the calcula-
tions because of signal overlap. The change in chemical shift upon
unfolding for Ala-CH3 was less then 0.03 ppm and was, therefore,
not investigated. The iterative curve fitting for the chemical shift of
Val-CHb did not converge. The first-order derivatives of the melting
curves for both diastereomeric compounds is available in the ESI.
k For diastereomer 7a, Dd/DT calculated between 213 and 273 K was
2.4 ppb Kꢁ1 for NHAla and 4.6 ppb Kꢁ1 for NHVal. For the diastereo-
mer 7b, the corresponding values were 5.6 (NHAla) and 6.9 ppb Kꢁ1
(NHVal). The corresponding values for solutions at or above room
temperature were considerably lower (Table 2).
838
New J. Chem., 2002, 26, 834–843

View Article Online
role of hydrophobic and hydrogen-bonding interactions to be
made. The folding of the model compound 7b in methanol
solution is exothermic, and the enthalpic (DH) and entropic
(ꢁTDS) contributions to its stability at room temperature
are approximately equal (Table 3). When changing the solvent
to DMSO, the folding still remains exothermic, but becomes
enthalpy driven. The increasing role of enthalpy upon chan-
ging the solvent indicates the greater influence of interstrand
hydrogen bonding on hairpin folding in the less polar aprotic
medium, in which, as expected, hydrophobic stabilization is
not as effective. The negative entropic contribution to folding
suggests that the hydrophobic effect is not the dominating sta-
bility-determining factor in our model system. As the folding
process becomes more exothermic (DH upon unfolding is more
positive in DMSO than in methanol, i.e. DH upon folding is
more negative), the entropy term gets increasingly more nega-
tive. Thus, the stronger hydrogen-bonding interactions are
partially compensated for by an adverse conformational
entropy term as the peptide backbone becomes more con-
strained in DMSO. However, the increase of the entropy term
(ꢁTDS) is overcompensated for by a larger decrease in the
enthalpic contribution (DH) when changing the solvent from
methanol to DMSO.
conformation of the (R,S)-diastereomer would be more prob-
able than that of the (S,S)-diastereomer (21% compared to
16%),the opposite of what is observed experimentally. The fail-
ure of the calculations might be attributable to overestimation
of the importance of steric repulsions.
Experimental
Materials
Starting materials were purchased from commercial suppliers
and were used without further purification. Diethylamine, di-
isopropylethylamine, (trimethylsilyl)acetylene, Pd(PPh3)2Cl2 ,
N,O-bis(trimethylsilyl)acetamide and 2-iodoaniline were
obtained from Aldrich. Methyl-2-iodobenzoate was from Lan-
caster, cuprous iodide from Merck, O-(7-azabenzotriazol-1-
yl)-N,N,N0,N0-tetramethyluronium hexafluorophosphate and
dimethylformamide were obtained from Fluka, hexane from
J. T. Baker, and L-valine and L-alanine methyl ester were pur-
chased from Bachem. Ethyl acetate was from Riedel-de Hae¨n.
Microwave heating
The method used is described in ref. 41.
The hydrophobic contribution to b-hairpin stability
Sonochemistry
Comparison of the thermodynamic parameters of the diaster-
eomeric compounds 7a and 7b in the same solvent provides an
opportunity to estimate the impact of the hydrophobic interac-
tions on hairpin stability. As described above, in methanol
solution, the high contribution of the electrostatic interaction
(i.e. hydrogen bonding) conceals the rather small hydrophobic
stabilizing effect. In the less competitive, polar solvent DMSO,
the extent of the hydrophobic effect can be estimated. The
observed peptide melting curves indicate that the close neigh-
borhood of apolar amino acid side chains provides significant
stability. Thus, the van der Waals interaction has a higher
influence on hairpin folding than the simultaneous, destabiliz-
ing steric crowding. In DMSO solution, a melting point for
diastereomer 7a more than 30 K higher than that of diastereo-
mer 7b indicates the extent of the influence of hydrophobic
interactions on b-hairpin population.
The sonochemical reactions were performed at ambient tem-
perature using a VCX 500 ultrasonic processor (Sonics and
Materials INC) equipped with a 13 mm diameter probe. 50%
Amplitude was selected for the irradiations.
HPLC
The separation of compound 7a and 7b was performed on a
Gilson HPLC system connected to a Dynamax 83-121 C col-
˚
umn (60 A) and a Dynamax absorbance detector (Model
UV-1) working at 254 nm.
Theoretical conformational analysis
The conformational energy calculations were performed using
the OPLS-AA all-atom force field as implemented in the pro-
gram Macromodel 7.0.42 The general born solvent accessible
(GB/SA) surface area method developed by Still43 was used
in all calculations. The number of torsion angles allowed to
vary during each Monte Carlo step ranged from 1 to n ꢁ 1,
where n is the total number of rotatable bonds. Amide bonds
were fixed in the trans configuration. Conformational searches
were conducted by use of the systematic unbound multiple
minimum (SUMM) search method44 implemented in the
Batchmin program. First, 20 000 Monte Carlo step runs were
performed (4000 steps on 5 processors with the command
NPRC, at the Centre for Parallel Computers, KTH, Stock-
holm) and those conformations within 25 kJ molꢁ1 of the glo-
bal minimum were kept. PR conjugate gradient minimization
with a maximum of 1000 iterations was used in the conforma-
tional search. In the subsequent minimization to fully con-
verged structures, a maximum of 50 000 steps of PRCG and/
or truncated Newton conjugated gradient (TNCG) minimiza-
tion was used.
Conclusions
The present investigation of the (S,S)- and (R,S)-diastereomers
of a hairpin mimetic allowed the qualitative evaluation of the
importance of intramolecular hydrogen bonding on b-hairpin
stability in several solvents to be carried out.
The population difference of the mimetics observed in sol-
vents possessing different polarity and ability to compete
(chloroform < dimethylsulfoxide < methanol) indicates the
critical role of hydrogen bonding in hairpin stability.
Comparison of the thermodynamic constants of the two dia-
stereomeric compounds in the same solvent allowed a quanti-
tative evaluation of the hydrophobic effect to be made. The
observed significantly higher thermodynamic stability of 7a
compared to 7b in DMSO solution suggests that the hydro-
phobic effect might have significant influence on hairpin stabi-
lity in polar solvents. The similar behavior of the diastereomers
in methanol solution indicates that in such a small model sys-
tem, the effect of a competing solvent dominates over weak
hydrophobic forces. For a complete understanding of the role
of hydrophobic effects on hairpin stability, a study involving
larger diastereomer model hairpins will be necessary. Investi-
gations in this direction are in progress.
Circular dichroism spectroscopy
CD spectra were measured on a JASCO J-810 spectropolari-
meter from 190 to 360 nm using a 0.2 mm path length cell. 5
Scans were accumulated at ambient temperature with a scan-
ning speed of 100 nm minꢁ1, using 50 mmol dmꢁ3 methanol
solutions. Optical rotation was measured with a Perkin-Elmer
241 polarimeter.
The experimental results could not be reproduced in every
respect by computational studies. The calculations performed
using the OPLS-AA force field predicted that the b-hairpin
New J. Chem., 2002, 26, 834–843
839

View Article Online
Infrared spectroscopy
prepared by acetylation of (S)-valine under sonochemical con-
ditions.45 (S)-Alaninemethylamide was obtained by transami-
dation of (S)-alanine methyl ester following literature
procedures.46 Subjecting 2-iodoaniline (1) to Sonogashira cou-
IR spectra were obtained on a Perkin Elemer 1600 series FTIR
instrument; recording 16 scans on 10 mol dmꢁ3 samples in
CHCl3 solution using a 1 mm cuvette.
pling with (trimethylsilyl)acetylene afforded 2-[(trimethylsily-
41
l)ethynyl]aniline (2) in excellent yield,
desilylation by
potassium fluoride47 yielded 2-ethynylaniline (3). Reaction of
compound 3 with methyl-2-iodobenzoate under the conditions
used in step c gave 2-amino-20-carboxymethyldiphenylacety-
lene (4). O-(7-Azabenzotriazol-1-yl)-N,N,N0,N0-teramethyluro-
nium hexafluorophosphate (HATU)-mediated coupling of 4
with (S)-N-acetylvaline yields 5. The amino group of 4 is steri-
cally and inductively deactivated, making this step unusually
difficult. Therefore, the amino group was activated with
N,O-bis(trimethylsilyl)acetamide (BSA) in dichloromethane
for 30 min. The carboxyl group was allowed to react for 20
min with HATU in a mixture of diisopropylethylamine,
dichloromethane, and dimethylformamide. The two solutions
were then combined and stirred for 72 h, giving 5 in 52% iso-
lated yield. Alternatively, the modified procedure of Roshchin
and Bumagin.48 gave 45% yield in 20 h. 5 was then hydrolysed
and racemized using four equivalents of potassium tert-butox-
ide in diethyl ether, affording 6 in quantitative yield. The
benzoic acid derivative 6 was then coupled to (S)-alanine-
methylamide using HATU as the coupling reagent, resulting
in a mixture of hairpin mimetic diastereomers 7a and 7b.
The mixture was purified by column chromatography on silica
using ethyl acetate as eluent, followed by separation of the two
diastereomers by preparative HPLC (ethyl acetate eluent).
This procedure allowed us to synthesize the two diastereomers
in parallel in an efficient manner. Two methods for the selec-
tive synthesis of 7a were earlier described by Kemp and
Li,16e but no yields were reported.
Mass spectrometry
Mass spectra (EI, 70 eV) were obtained with a Hewlett Pack-
ard 5971 Series mass selective detector interfaced with a Hew-
lett Packard 5890 Series II gas chromatograph equipped with a
DB-1 (25 m ꢄ 0.20 mm) capillary column. The ESI-MS of
compound 7a and 7b was obtained with a Finnigan Thermo-
Quest AQA mass spectrometer (ESI 30 eV, probe temperature
100 ꢂC) equipped with a Gilson 322-H2 gradient pump system
and a SB-C18 column. A water–acetonitrile–formic acid
(0.05%) mobile phase was used with a gradient of 20 to 80%
acetonitrile over 10 min.
NMR spectroscopy
NMR spectra were recorded on a Jeol JNM EX400 spectro-
meter (1H at 399.8 MHz, 13C at 100.5 MHz), a Varian UNITY
spectrometer (1H at 399.95 MHz, 13C at 100.6 MHz), and on a
JEOL JNM EX270 spectrometer (1H at 270.2 MHz, 13C at
67.8 MHz). Chemical shifts are referenced indirectly to TMS
via the 2H lock signal. Exchange of amide protons with deuter-
ons was affected by dissolving the sample (ca. 10 mg) in a 1:1:1
mixture of D2O–CD3OD–acetone-d6 (1 mL), followed by eva-
poration of the solvent after 3 h. NOE effects were measured
from NOESY and ROESY spectra with mixing times between
0.7 and 1.5 s. In the 1D NOE difference experiments, a satura-
tion time of 20 s was used.
(S)-2-Acetylamino-3-methylbutyric acid. (S)-Valine (470.6
mg, 4 mmol) was dissolved in water (10 mL) and was sonicated
for 6 min.45 Acetic acid anhydride (0.75 mL, 8 mmol) was
added at 0, 2, and 4 min. Then, the mixture was concentrated
on a rotatory evaporator. The residue was dissolved in metha-
nol and the solution filtered, concentrated on a rotatory eva-
porator, and the remaining solvent removed under reduced
pressure overnight, giving (S)-2-acetylamino-3-methylbutyric
Thermodynamic analysis
The NMR melting curve of each proton investigated was first
analysed using reciprocal temperature plots and its first-order
derivative. The thermodynamic parameters Tm and DHm esti-
mated in this way (eqn. 1 and 2) were used as initial conditions
in the restricted least-squares calculations fitting of the experi-
mental data to eqn. 3xx using Scientist (Windows version 1.04,
MicroMath Inc., Salt Lake City, UT, USA). Method details
are described by Honda and co-workers.40a,b
1
acid as a white solid in 99% yield (629.5 mg, 4.0 mmol). H
3
NMR (400 MHz, D2O, 25 ꢂC): d 4.17 [d, J(H,H) ¼ 5.9 Hz,
1H, CHa], 2.11 [dh, 3J(H,H) ¼ 4.8, 5.9 Hz, 1H, CHb], 1.98
3
(s, 3H, COCH3), 0.89 [2d, J(H,H) ¼ 4.8 Hz, 6H, CH3g]. 13C
dobs ¼ dF þ ðdU ꢁ dFÞf
ð1Þ
ð2Þ
NMR (100 MHz, D2O, 25 ꢂC) d 175.7, 174.5, 58.6, 29.8,
21.6, 18.3, 17.2.
2
3
5
4R
dd
4
ꢀ
ꢁ
DHm
¼
þ
dF ꢁ dU
1
d
=
T
(S)-2-Amino-N-methyl-propionamide. (S)-Alanine methyl
ester hydrochloride (1.00 g, 7.2 mmol) was treated with 40%
methylamine in water (60 mL) following the literature proce-
dure.46 The solution was stirred at 40 ꢂC for 80 min, and then
concentrated under reduced pressure. The residue was dis-
solved in ethanol (253 K) and treated with diethyl ether (253
K) until a white precipitate was observed. The precipitate
(methylamine salt) was filtered off and the filtrate was concen-
trated on a rotatory evaporator, followed by removal of the
remaining solvent under reduced pressure, giving 2-amino-N-
methylpropionamide as a white solid in 84% yield (611.0 mg,
m
dF ꢁ dU
h
ꢀ
ꢁi
dobs ¼ dU
ð3Þ
ꢁDHm
1
1
m
1 þ exp
T ꢁ T
R
X-Ray crystallography
Details of the X-ray crystallographic analysis of the structure
of 7a are given in the ESI.
CCDC reference number 183993. See http://www.rsc.org/
suppdata/nj/b1/b111241d/ for crystallographic data in CIF
or other electronic format.
1
6.0 mmol). H NMR (400 MHz, D2O, 25 ꢂC): d 3.58 [q,
3J(H,H) ¼ 6.9 Hz, 1H; CHa], 2.71 (s, 3H; NH–CH3), 1.27
3
[d, J(H,H) ¼ 6.9 Hz, 3H; CH3]. 13C NMR (100 MHz, D2O,
25 ꢂC) d 176, 49.9, 25.9, 19.0.
Synthesis
Scheme 1 outlines the methodology used to prepare the (S)-
Val, (S)-Ala-derivative (7a) and the (R)-Val, (S)-Ala derivative
(7b) of 2-amido-20-carboxamidotolane. (S)-N-Acetylvaline was
2-[(Trimethylsilyl)ethynyl]aniline (2). 2-Iodoaniline (1) (394
mg, 1.8 mmol), Pd(PPh3)2Cl2 (25.6 mg, 0.04 mmol), CuI
(13.8 mg, 0.08 mmol), (trimethylsilyl)acetylene (0.28 mL,
2.00 mmol), diethylamine (3.0 mL, 27.20 mmol), and dimethyl-
formamide (0.5 mL) were stirred under nitrogen in a heavy-
xx The original reference40b contains an erroneous version of eqn. 3.
840
New J. Chem., 2002, 26, 834–843

View Article Online
walled Smith process vial at 120 ꢂC for 5 min under microwave
irradiation. Then, the reaction mixture was poured into 0.1 M
aqueous HCl (5–10 mL) and extracted three times with diethyl
ether (5–10 mL). The combined organic layers were washed
with conc. aqueous NaHCO3 solution (5–10 mL) and water
(5–10 mL), re-extracting the aqueous phases twice in each case
with diethyl ether, then concentrated under reduced pressure.
Thereafter, the residue was purified by flash chromatography
(silica gel 60, particle size 0.040–0.063 mm, Merck; hexane–
ethyl acetate 12:1). The combined product fractions were con-
centrated on a rotatory evaporator, followed by removal of the
remaining solvent under reduced pressure overnight, yielding 2
as a brown oil in 98% yield (331.8 mg, 1.76 mmol).41 1H NMR
(400 MHz, CDCl3 , 25 ꢂC): d 7.29 [dd, J(H,H) ¼ 1.7, 7.5 Hz,
1H; ArH], 7.11 [ddd, J(H,H) ¼ 1.7, 7.5, 8.2 Hz, 1H; ArH],
6.68 [dd, J(H,H) ¼ 1.1, 8.2 Hz, 1H; ArH], 6.66 [ddd,
J(H,H) ¼ 1.1, 7.5, 7.5 Hz, 1H; ArH], 4.07 (br s, 2H; NH2),
0.27 [s, 9H; RSi(CH3)3]. 13C NMR (100 MHz, CDCl3 ,
25 ꢂC) d 148.2, 132.2, 129.8, 117.7, 114.1, 107.7, 101.7, 99.8,
0.0. MS (70 eV, EI): m/z (%) 189 (61) [M]+, 174 (100). IR
(CHCl3): n 3617.2, 3474.8, 3383.7, 3019.1, 2967.8, 2893.8,
2H; NH2), 3.94 (s, 3H; CH3). 13C NMR (100 MHz, CDCl3 ;
25 ꢂC) d 166.2, 149.6, 133.7, 132.0, 131.9, 130.5, 130.2, 130.2,
127.3, 124.7, 117.1, 114.1, 107.2, 93.6, 92.5, 52.3. MS (70 eV,
EI): m/z (%) 251 (83) [M]+, 219 (100), 190 (44). IR (CHCl3)
n 3690.3, 3489.1, 3369.5, 3023.5, 2395.6, 2207.7, 1716.5,
1623.2, 1491.6 cmꢁ1
.
(S)-2-[20-(200-Acetylamino-300-methylbutyrylamino)pheny-
lethynyl]benzoic acid methyl ester (5). Compound 4 (60.7 mg,
0.24 mmol) dissolved in dichloromethane (1 mL) was treated
with BSA (0.06 mL, 0.24 mmol) for 30 min. 2-Acetylamino-
3-methylbutyric acid was dissolved in a mixture of dichloro-
methane (2 mL), dimethylformamide (1 mL), and diisopropy-
lethylamine (0.21 mL, 1.21 mmol). HATU (286.6 mg, 1.21
mmol) was added and the mixture was stirred for 20 min.
The two solutions were combined and the mixture was stirred
for 72 h. Thereafter, the reaction mixture was poured into 0.1
M aqueous HCl (5–10 mL) and extracted three times with
diethyl ether (5–10 mL). The combined organic layers were
washed with conc. aqueous NaHCO3 solution (5–10 mL)
and water (5–10 mL), re-extracting the aqueous phases twice
in each case with diethyl ether, filtered through magnesium sul-
fate, and then concentrated under reduced pressure. The resi-
due was purified by flash chromatography (silica gel 60,
particle size 0.040–0.063 mm, Merck; hexane–ethyl acetate
1:1). The combined product fractions were concentrated on a
rotatory evaporator, followed by removal of the remaining sol-
vent under reduced pressure overnight, giving 5 as white solid
in 52% yield (49.1 mg 0.13 mmol). Alternatively, the modified
procedure of Roshchin and Bumagin.48 gave 45% yield in 20 h.
1H NMR (400 MHz, CDCl3 , 25 ꢂC): d 9.50 (br s, 1H; NH),
8.55 [dd, J(H,H) ¼ 1.2, 8.2 Hz, 1H; ArH], 8.12 [ddd,
J(H,H) ¼ 0.6, 1.4, 8.3 Hz, 1H; ArH], 7.68 [ddd, J(H,H) ¼
0.6, 1.4, 7.8 Hz, 1H; ArH], 7.57 [ddd, J(H,H) ¼ 1.40, 7.5,
7.8 Hz, 1H; ArH], 7.53 [ddd, J(H,H) ¼ 1.6, 7.1, 8.3 Hz, 1H;
ArH], 7.42 [dd, J(H,H) ¼ 1.4, 7.4 Hz, 1H; ArH], 7.36 [ddd,
J(H,H) ¼ 1.1, 1.4, 8.3 Hz, 1H; ArH], 7.09 [ddd,
2141.8, 1611.5, 1490.6, 1454.3 cmꢁ1
.
2-Ethynylaniline (3). Following the literature procedure,47
compound 2 (331.8 mg, 1.76 mmol) was dissolved in methanol
(10 mL) and KFꢀ2H2O (495.0 mg, 5.3 mmol) was added. The
mixture was stirred for 7 h at room temperature and thereafter
concentrated under reduced pressure. The residue was dis-
solved in diethyl ether and water was added. The organic phase
was separated, and the aqueous phase was re-extracted three
times. The combined organic phases were concentrated under
reduced pressure, giving compound 3 as a yellowish solid in
97% yield (198.8 mg, 1.7 mmol). 1H NMR (400 MHz, CDCl3 ,
25 ꢂC): d 7.32 [dd, J(H,H) ¼ 1.6, 7.6 Hz, 1H; ArH], 7.15 [ddd,
J(H,H) ¼ 1.6, 7.4, 7.4 Hz, 1H; ArH], 6.70 [dd, J(H,H) ¼ 0.9,
7.4 Hz, 1H; ArH], 6.68 [ddd, J(H,H) ¼ 0.9, 7.4, 7.6 Hz, 1H;
ArH], 4.25 (br s, 2H; NH2). 13C NMR (100 MHz, CDCl3 ,
25 ꢂC) d 148.6, 132.7, 130.2, 117.9, 114.4, 106.7, 82.5, 80.7.
MS (70 eV, EI): m/z (%) 117 (100) [M]+, 90 (60). IR (CHCl3)
n 3496.5, 3399.6, 3302.7, 2927.1, 2854.4, 2091.2, 1609.5, 1488.4,
3
J(H,H) ¼ 1.4, 7.6, 7.6 Hz, 1H; ArH], 6.48 [d, J(H,H) ¼ 7.9
Hz, 1H; NH], 5.30 [dd, 3J(H,H) ¼ 5.9, 7.9 Hz, 1H; CHa],
4.14 (s, 3H; CH3), 2.20 [dh, 3J(H,H) ¼ 5.9, 6.8 Hz, 1H;
CHb], 2.04 (s, 3H; CH3), 1.01 [d, 3J(H,H) ¼ 6.8 Hz, 3H;
CH3], 0.96 [d, 3J(H,H) ¼ 6.8 Hz, 3H; CH3]. 13C NMR (100
MHz, CDCl3 , 25 ꢂC) d 171.2, 169.7, 166.3, 140.1, 133.8,
132.4, 132.0, 130.8, 130.0, 129.9, 128.2, 124.0, 123.4, 119.7,
112.2, 95.7, 90.4, 57.7, 53.4, 32.8, 23.4, 19.2, 17.8. MS (70
eV, EI): m/z (%) 392 (6) [M]+, 281 (9), 251 (100), 207 (58),
72 (76). IR (CHCl3) n 3422.5, 3322.2, 2400.0, 1716.0, 1666.0,
1458.1 cmꢁ1
.
2-(20-Aminophenylethynyl)benzoic acid methyl ester (4).
Compound 3 (200.2 mg, 1.71 mmol), 2-iodobenzoic acid
methyl ester (493.0 mg, 1.88 mmol), Pd(PPh3)2Cl2 (24.0 mg,
0.034 mmol), CuI (13.0 mg, 0.068 mmol), diethylamine (1.5
mL, 13.60 mmol), and dimethylformamide (0.5 mL) were stir-
red under nitrogen in a heavy-walled Smith process vial at
120 ꢂC for 5 min under microwave irradiation. Then, the reac-
tion mixture was filtered through Celite and was poured into
0.1 M aqueous HCl (5–10 mL). The mixture was extracted
three times with diethyl ether (5–10 mL). The combined
organic layers were washed with conc. aqueous NaHCO3 solu-
tion (5–10 mL) and water (5–10 mL), re-extracting the aqueous
phases twice in each case with diethyl ether, filtered through
magnesium sulfate, and then concentrated under reduced pres-
sure. Thereafter, the residue was purified by flash chromato-
graphy (silica gel 60, particle size 0.040–0.063 mm, Merck;
hexane–ethyl acetate 9:1). The combined product fractions
were concentrated on a rotatory evaporator, followed by
removal of the remaining solvent under reduced pressure over-
night, giving 4 as a white solid in 78% yield (336.3 mg 1.33
1578.5, 1525.4, 1509.8, 1447.3 cmꢁ1
.
(S)-2-[20-(200-Acetylamino-300-methylbutyrylamino)pheny-
lethynyl]benzoic acid (6). Compound 5 (390.9 mg, 1.0 mmol)
was dissolved in diethyl ether (5 mL) and dichloromethane
(1 mL). A suspension of potassium tert-butoxide (446.7 mg,
4.0 mmol) in diethyl ether (20 mL) was added to this solution
and the mixture was stirred for 15 h. Then, the suspension was
extracted with chloroform. The combined organic layers were
filtered through magnesium sulfate, then concentrated on a
rotatory evaporator, followed by removal of the remaining sol-
vent under reduced pressure, giving 5 as brown solid in 99%
yield (373.0 mg, 0.98 mmol). 1H NMR (270 MHz, CDCl3 ,
25 ꢂC): d 9.57 (br s, 1H; NH), 8.47 [d, 3J(H,H) ¼ 8.6 Hz,
1H; ArH], 8.09 [d, 3J(H,H) ¼ 7.3 Hz, 1H; ArH], 7.66 [d,
3J(H,H) ¼ 7.6 Hz, 1H; ArH], 7.51–7.56 (m, 2H; ArH), 7.32–
1
mmol). H NMR (400 MHz, CDCl3 , 25 ꢂC): d 8.02 [ddd,
J(H,H) ¼ 0.6, 1.36, 8.0 Hz, 1H; ArH], 7.67 [ddd, J(H,H) ¼
0.6, 1.4, 7.8 Hz, 1H; ArH], 7.51 [ddd, J(H,H) ¼ 1.36, 7.48,
7.80 Hz, 1H; ArH], 7.40 [ddd, J(H,H) ¼ 1.40, 7.48, 7.90 Hz,
1H; ArH], 7.30 [ddd, J(H,H) ¼ 0.56, 1.56, 7.67 Hz, 1H;
ArH], 7.15 [ddd, J(H,H) ¼ 1.56, 7.32, 8.11 Hz, 1H; ArH],
6.74 [ddd, J(H,H) ¼ 0.56, 1.12, 8.11 Hz, 1H; ArH], 6.68
[ddd, J(H,H) ¼ 1.12, 7.67, 7.32 Hz, 1H; ArH], 5.17 (br s,
3
7.44 (m, 2H; ArH), 7.09 [t, J ¼ 7.6 Hz, 1H; ArH], 6.90 (bs,
1H; NH), 5.30 [dd, 3J(H,H) ¼ 4.3, 7.3 Hz, 1H; CHa], 2.10
(s, 3H; CH3), 1.22 [dh, 3J(H,H) ¼ 5.9, 7.6 Hz, 1H; CHb],
0.99 [2d, 3J(H,H) ¼ 5.9 Hz, 6H; CH3]. 13C (67.9 MHz, CDCl3 ,
25 ꢂC) d 170.9, 170.88, 168.4, 139.9, 133.1, 132.1, 132.0, 131.3,
130.6, 129.9, 128.2, 123.6, 119.7, 112.5, 95.2, 90.0, 58.5, 33.2,
New J. Chem., 2002, 26, 834–843
841

View Article Online
31.1, 23.6, 19.2, 18.2. IR (CHCl3) n 3412.2, 3309.7, 2400.0,
1698.0, 1681.0, 1659.6, 1582.7, 1522.9, 1450.3 cmꢁ1
Acknowledgements
.
The Centre for Parallel Computers (PDC) at the Royal Insti-
tute of Technology (KTH), Stockholm, Sweden, is acknowl-
edged for a generous allotment of computing time. We thank
G. Lindeberg, A. Johansson, and E. Akerblom for helpful dis-
cussions.
2-[20-(200-Acetylamino-300-methylbutyrylamino)phenylethynyl]-
N-(1-methylcarbamoylethyl)benzamide (7). Compound 6 (373.0
mg, 1.0 mmol) was dissolved in dichloromethane (14 mL), di-
isopropylethylamine (1.57 mL, 0.12 mol) and HATU (943.3
mg, 4.0 mmol) was added, and the mixture was stirred for 1
h. 2-Amino-N-methylpropionamide (411.5 mg, 4.0 mmol)
dissolved in a mixture of dimethylformamide (3 mL), dichloro-
methane (1 mL), and diisopropylethylamine (0.1 mL) was
added and the solution was stirred for 140 min. Then, the reac-
tion mixture was poured into 0.1 M aqueous HCl (5–10 mL).
The mixture was extracted three times with diethyl ether (5–10
mL). The combined organic layers were washed with conc.
aqueous NaHCO3 solution (5–10 mL) and water (5–10 mL),
re-extracting the aqueous phases twice in each case with
diethyl ether, filtered through magnesium sulfate, and then
concentrated under reduced pressure. Thereafter, the residue
was purified by flash chromatography (silica gel 60, particle
size 0.040–0.063 mm, Merck; ethyl acetate). The combined
product fractions were concentrated on a rotatory evaporator.
The diastereomeric mixture was separated by HPLC using
ethyl acetate as eluent, giving 129.1 mg of compound 7a and
107.1 mg of 7b, 51% yield (236.2 mg, 0.1 mmol).
˚
References
1
2
N. C. J. Strynadka, S. E. Jensen, P. M. Alzari and M. N. G.
James, Nat. Struct. Biol., 1996, 3, 290.
(a) J. D. Puglishi, L. Chen, S. Blanchard and A. D. Frankel,
Science, 1995, 270, 1200; (b) G. Variani Acc. Chem. Res., 1997,
30, 189.
3
4
(a) P. E. Floreancing, S. E. Swalley, J. W. Trauger and P. B. Der-
van, J. Am. Chem. Soc., 2000, 122, 6342; (b) R. E. Bremer, J. W.
Szewczyk, E. E. Baird and P. B. Dervan, Bioorg. Med. Chem.,
2000, 8, 1947–1955.
(a) J. Collinge Annu. Rev. Neurosci., 2001, 24, 519; (b) M. Mihara,
Y. Takahashi and A. Ueno, Biopolymers, 1998, 47, 83; (c) K. M.
Pan, M. Baldwin, J. Nguyen, M. Gasset, A. Serban, D. Groth, I.
Mehlhorn, Z. Huang, R. J. Fletterick, F. E. Cohen and S. B. Pru-
siner, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 10 962; (d ) R.
Riek, S. Hornemann, G. Wider, M. Billeter, R. Glockshuber
and K. Wuthrich, Nature, 1996, 382, 180.
5
(a) L. C. Serpell, C. C. F. Blake and P. E. Fraser, Biochemistry,
2000, 39, 13 269; (b) S. Ohnisiki, A. Koide and S. Koide, J.
Mol. Biol., 2000, 301, 477; (c) D. R. Booth, M. Sunde, V. Bellotti,
C. V. Robinson, W. L. Hutchinson, P. E. Fraser, P. N. Hawkins,
C. M. Dodson, S. E. Radford, C. C. F. Blake and B. M. Pepys,
Nature, 1997, 385, 787.
(S,S)-2-[20-(200-Acetylamino-300-methylbutyrylamino)pheny-
lethynyl]-N-(1-methylcarbamoylethyl)benzamide (7a). 1H
NMR (400 MHz, CDCl3 , 25 ꢂC): d 9.33 (br s, 1H; NH), 8.61
6
7
J. M. McDonnell, D. Fushman, S. M. Cahill, B. J. Sutton and D.
Cowburn, J. Am. Chem. Soc., 1997, 119, 5321.
3
3
[d, J(H,H) ¼ 8.4 Hz, 1H; ArH], 7.91 [d, J(H,H) ¼ 4.7 Hz,
1H; NH], 7.63–7.68 (m, 2H; ArH), 7.46–7.53 (m, 2H; ArH),
7.33–7.40 (m, 2H; ArH), 7.14 [d, 3J(H,H) ¼ 8.1 Hz; NH],
7.06 [dd, 3J(H,H) ¼ 7.5, 7.5 Hz, 1H; ArH], 6.60 [d,
3J(H,H) ¼ 9.3 Hz, 1H; NH], 5.50 [dd, 3J(H,H) ¼ 5.2, 9.3
(a) J. P. Derrick, G. J. Davies, Z. Dauter, K. S. Wilson and D. B.
Wigley, J. Mol. Biol., 1992, 227, 1253; (b) J. P. Derrick and D. B.
Wigley, Nature, 1992, 359, 752; (c) J. P. Derrick and D. B. Wigley,
J. Mol. Biol., 1994, 243, 906; (d ) C. Boutillon, R. Wintjens, G.
Lippens, H. Drobecq and A. Tartar, Eur. J. Biochem., 1995,
231, 166; (e) E. L. McCallister, E. Alm and D. Baker, Nat. Struct.
Biol., 2000, 7, 669.
3
Hz, 1H; Val-CHa], 5.32 [dq, J(H,H) ¼ 6.8, 8.4 Hz, 1H; Ala-
CHa], 2.76 [d, 3J(H,H) ¼ 4.7 Hz, 3H; CH3NHAla], 2.33 [dh,
3J(H,H) ¼ 5.2, 6.6 Hz, 1H; Val-CHb], 2.1 (s, 3H, CH3COVal),
1.49 [d, 3J(H,H) ¼ 6.8 Hz, 3H; Ala-CH3], 1.05 [d,
3J(H,H) ¼ 6.6 Hz, 3H; Val-CH3g], 0.95 [d, 3J(H,H) ¼ 6.6
8
9
(a) P. D. Kwong, R. Wyatt, J. Robinson, R. W. Sweet, J. Sodroski
and W. A. Hendrickson, Nature, 1998, 393, 648; (b) A. Cochran,
R. T. Tong, M. A. Starovanski, E. J. Park, R. S. McDowell, J. E.
Theaker and N. J. Skelton, J. Am. Chem. Soc., 2001, 123, 625.
(a) B. Odaert, F. Jean, C. Boutillon, E. Buisine, O. Melnyk, A.
Tartar and G. Lippens, Protein Sci., 1999, 8, 2773; (b) C. Das,
S. Raghothama and P. Balaram, J. Am. Chem. Soc., 1998, 120,
13
Hz, 3H; Val-CH3g]. C NMR (100 MHz, CDCl3 , 25 ꢂC) d
173.5,171.3, 170.6, 166.5, 140.3, 135.2, 133.9, 132.3, 130.9,
129.8, 128.3, 127.3, 123.4, 122.3, 119.4, 112.1, 94.9, 89.5,
57.7, 48.5, 32.8, 26.0, 23.7, 20.4, 19.2, 17.2. MS (30 eV, ESI):
m/z (%) 463 (100) [M ꢁ H]+. IR (CHCl3) n 3412.9, 3307.1,
´
5812; (c) E. Alba, A. Jimenez and M. Rico, J. Am. Chem. Soc.,
1997, 119, 175.
10 (a) H. J. Dyson and P. E. Wright, Annu. Rev. Biophys. Chem.,
1991, 20, 519; (b) V. Munoz and L. Serrano, Proteins, 1994, 20,
301; (c) T. P. Quinn, N. B. Tweedy, R. W. Williams, J. S. Richard-
son and D. C. Richardson, Proc. Natl. Acad. Sci. U. S. A., 1994,
91, 8747.
11 (a) B. L. Sibanda and J. M. Thornton, J. Mol. Biol., 1993, 229,
428; (b) B. L. Sibanda and J. M. Thornton, Nature, 1985, 316,
170
2972.5,
2935.8,
2876.4,
1648.7,
1578.9,
1514.2
cmꢁ1.[a]2D0 ¼ +64ꢂ (c ¼ 1, CHCl3).
(R,S)-2-[20-(200-Acetylamino-300-methylbutyrylamino)phenyl-
ethynyl]-N-(1-methylcarbamoylethyl)benzamide (7b). 1H
NMR (400 MHz, CDCl3 , 25 ꢂC): d 9.68 (br s, 1H; NH), 8.63
3
3
[d, J(H,H) ¼ 8.4 Hz, 1H; ArH], 7.89 [d, J(H,H) ¼ 4.7 Hz,
1H; NH], 7.71 [d, 3J(H,H) ¼ 7.7 Hz, 1H; ArH], 7.65 [d,
3J(H,H) ¼ 7.5 Hz, 1H; ArH], 7.45–7.52 (m, 2H; ArH), 7.30–
7.42 (m, 2H; ArH), 7.06 [t,3J(H,H) ¼ 7.5 Hz, 1H; ArH], 6.96
[d, 3J(H,H) ¼ 9.0 Hz; NH], 6.68 [d, 3J(H,H) ¼ 9.1 Hz, 1H;
NH], 5.65 [dd, 3J(H,H) ¼ 4.4, 9.1 Hz, 1H; Val-CHa], 5.26
[dq,3J(H,H) ¼ 7.0, 9.0 Hz, 1H; Ala-CHa], 2.81 [d,
3J(H,H) ¼ 4.7 Hz, 3H; CH3NHAla], 2.25 [dh,3J ¼ 4.4, 6.8 Hz,
12 For reviews on b-hairpin mimetics, see: (a) M. S. Searle J. Chem.
Soc., Perkin Trans. 2, 2001, 1011; (b) J. A. Robinson Synlett,
1999, 4, 429; (c) J. S. Nowick, E. M. Smith and M. Pairish, Chem.
Soc. Rev., 1996, 401.
13 (a) D. L. Minor, Jr. and S. P. Kim, Nature, 1994, 367, 660; (b)
C. K. Smith, J. M. Withka and L. Regan, Biochemistry, 1994,
33, 5510; (c) P. Y. Chou and G. D. Fasman, Biochemistry,
1973, 13, 211; (d ) K. Smith and L. Regan, Acc. Chem. Res.,
1997, 30, 153; (e) K. Gunasekaran, C. Ramakrishnan and P.
Balaram, Protein Eng., 1997, 10, 1131.
1H; Val-CHb], 2.1 (s, 3H; CH3COVal), 1.36 [d, J(H,H) ¼ 7.0
3
Hz, 3H; Ala-CH3], 0.94 [d, 3J(H,H) ¼ 6.8 Hz, 3H; Val-
14 S. J. Russel and A. G. Cochran, J. Am. Chem. Soc., 2000, 123,
625.
3
CH3g], 0.87 [d, J(H,H) ¼ 6.8 Hz, 3H; Val-CH3g]. 13C NMR
15 (a) I. Pozdnyakova, J. Guidry and P. Wittung-Stafshede, J. Am.
Chem. Soc., 2000, 122, 6337; (b) A. C. Gibbs, L. H. Kondejewski,
W. Gronwald, A. M. Nip, R. S. Hodges, B. D. Sykes and D. S.
Wishart, Nat. Struct. Biol., 1998, 5, 284; (c) D. Tolkatchev, A.
Ng, W. Vranken and F. Ni, Biochemistry, 2000, 39, 2878.
16 (a) M. J. Soth and J. S. Nowick, J. Org. Chem., 1999, 64, 276; (b)
S. C. Shankaramma, S. K. Singh, A. Sathyamurthy and P.
Balaram, J. Am. Chem. Soc., 1999, 121, 5360; (c) T. S. Haque,
J. C. Little and S. H. Gellman, J. Am. Chem. Soc., 1996, 118,
(100 MHz, CDCl3 , 25 ꢂC) d 173.6, 171.4, 170.3, 166.3, 140.8,
135.6, 133.3, 132.0, 130.8, 129.8, 128.3, 127.0, 123.3, 122.3,
119.3, 112.0, 94.6, 89.6, 57.3, 48.5, 33.2, 26.0, 23.7, 19.9,
19.3, 17.0. MS (30 eV, ESI): m/z (%) 463 (46) [M ꢁ H]+, 432
(24), 322 (100), 291 (9), 263 (7). IR (CHCl3) n 3413.3, 3307.6,
2973.3, 2933.5, 2875.2, 1653.0, 1580.0, 1517.7 cmꢁ1
[a]2D0 ¼ ꢁ64ꢂ (c ¼ 1, CHCl3).
.
842
New J. Chem., 2002, 26, 834–843

View Article Online
6975; (d ) Y. J. Chung, B. R. Huck, L. A. Christianson, H. E.
Stanger, S. Krautha¨user, D. R. Powell and S. H. Gellman, J.
Am. Chem. Soc., 2000, 122, 3995; (e) D. S. Kemp and Z. Q. Li,
Tetrahedron Lett., 1995, 36, 4175; ( f ) D. S. Kemp and Z. Q. Li,
Tetrahedron Lett., 1995, 36, 4179; (g) J. S. Nowick and F. Insaf,
J. Am. Chem. Soc., 1997, 119, 10 903; (h) J. S. Nowick Acc. Chem.
Res., 1999, 32, 287; (i) P. Chitnumsub, W. R. Fiori, H. A. Lashuel,
H. Diaz and J. W. Kelly, Bioorg. Med. Chem., 1999, 7, 39.
17 (a) I. L. Karle, S. K. Awasthi and P. Balaram, Proc. Natl. Acad.
Sci. U. S. A., 1996, 93, 8189; (b) B. L. Sibanda, T. L. Blundell and
J. M. Thornton, J. Mol. Biol., 1989, 206, 759.
18 T. S. Haque, J. C. Little and S. H. Gellman, J. Am. Chem. Soc.,
1994, 116, 4105–4106.
19 (a) D. L. Minor, Jr. and P. S. Kim, Nature, 1994, 371, 264; (b) F.
Xu and A. Cross, Magn. Reson. Chem., 1998, 36, 651.
20 A. G. Street and S. L. Mayo, Proc. Natl. Acad. Sci. U. S. A., 1999,
96, 9074.
21 (a) K. L. Constantine, L. Mueller, N. H. Andersen, H. Tong,
C. F. Wandler, M. S. Friedrichs and R. E. Bruccoleri, J. Am.
Chem. Soc., 1995, 117, 10 841; (b) M. S. Friadrichs, T. R. Stouch,
R. E. Bruccoleri, L. Mueller and K. L. Constantine, J. Am. Chem.
Soc., 1995, 117, 10 855.
22 (a) G. P. Dado and S. H. Gellman, J. Am. Chem. Soc., 1993, 115,
4228; (b) E. Gallo and S. H. Gellman, J. Am. Chem. Soc., 1993,
115, 9774; (c) E. Gallo and S. H. Gellman, J. Am. Chem. Soc.,
1994, 116, 11 560.
23 (a) K. Y. Tsang, H. Diaz, N. Graciani and J. W. Kelly, J. Am.
Chem. Soc., 1994, 116, 3988; (b) H. Wang and S. S. Sung, J.
Am. Chem. Soc., 2000, 122, 1999; (c) A. J. Maynard, G. J. Shar-
man and M. S. Searle, J. Am. Chem. Soc., 1998, 120, 1996.
24 T. S. Haque and S. H. Gellman, J. Am. Chem. Soc., 1997, 119,
2303.
su,R.,Isobe, H. Hanazono, Y. Soeda, M. Inagaki, N. Ito, R.
Higuchi and M. Yamamoto, Chem. Pharm. Bull., 1999, 47, 1040.
32 (a) H. Tsunematsu, S. Hachiyama,R.,Isobe, E. Ishida, M. Kakoi
and M. Yamamoto, Biochem. Biophys. Res. Commun., 1987, 146,
907; (b) H. Tsunematsu, S. Nakashima and M. Yamamoto, Org.
Mass Spectrom., 1989, 24, 943.
33 (a) A. Wokaun and R. R. Ernst, Chem. Phys. Lett., 1977, 52, 407;
(b) A. J. Shaka and R. Freeman, J. Magn. Reson., 1983, 51, 169.
34 M. Perpick-Dumont, W. F. Reynolds and R. G. Enriquez, Magn.
Reson. Chem., 1988, 26, 358.
35 (a) H. Kessler, C. Griesinger, J. Zarbock and H. R. Loosli, J.
Magn. Reson., 1984, 57, 331; (b) H. Kessler, C. Griesinger and
J. Lautz, Angew. Chem., Int. Ed. Engl., 1984, 23, 444; Angew.
Chem., 1984, 96, 434; (c) H. Kessler, C. Griesinger and K.
Wagner, J. Am. Chem. Soc., 1987, 109, 6927.
36 (a) A. Kumar, R. R. Ernst and K. Wuthrich, Biochem. Biophys.
Res. Commun., 1980, 95, 1; (b) G. Bodenhausen, H. Kogler and
R. R. Ernst, J. Magn. Reson., 1984, 58, 370.
37 A. Bax and D. G. Davis, J. Magn. Reson., 1985, 63, 207.
38 T. Parella Magn. Reson. Chem., 1998, 36, 467.
39 J. K. M. Sanders and J. D. Mersh, in Progress in Nuclear Mag-
netic Resonance Spectroscopy, ed. J. W. Emsley, J. Feeney and
L. H. Sutcliffe, Pergamon Press, Oxford, 1982, vol. 15, pp. 353.
40 (a) S. Honda, N. Kobayashi and E. Munekata, J. Mol. Biol.,
2000, 295, 269; (b) N. Kobayashi,S.,Honda, H. Yoshii and E.
Munekata, Biochemistry, 2000, 39, 6564; (c) S. Honda, N.
Kobayashi, E. Munekata and H. Uedaira, Biochemistry, 1999,
38, 1203.
´
41 M. Erdelyi and A. Gogoll, J. Org. Chem., 2001, 66, 4165.
42 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson and W. C. Still, J.
Comput. Chem., 1990, 11, 440.
25 M. S. Searle, S. R. Griffith-Jones and H. Skinner-Smith, J. Am.
Chem. Soc., 1999, 121, 11 615.
43 W. C. Still, A. Tempczyk, R. C. Hawley and T. Hendrickson, J.
Am. Chem. Soc., 1990, 112, 6127.
26 G. J. Sharman and M. S. Searle, J. Am. Chem. Soc., 1998, 120,
5291.
44 J. M. Goodman and W. C. Still, J. Comput. Chem., 1991, 12,
1110.
27 (a) L. Belvisi, C. Gennari, A. Mielgo, D. Potenza and C. Scolas-
tico, Eur. J. Org. Chem., 1999, 389; (b) R. R. Gardner, G. B.
Liang and S. H. Gellman, J. Am. Chem. Soc., 1999, 121, 1806.
28 L. Halab and W. D. Lubell, J. Org. Chem., 1999, 64, 3312.
29 D. Ranganathan, V. Haridas, S. Kurur, R. Nagaraj, E. Bikshapa-
thy, A. C. Kunwar, A. V. S. Sarma and M. Vairamani, J. Org.
Chem., 2000, 65, 365.
45 M. Pavlik, I. Kluh, F. Pavlikova, S. Vasickova and V. Kostka,
Collect. Czech. Chem. Commun., 1989, 1940.
46 R. W. Feenestra, E. H. M. Stokkingreef, A. M. Reichwein, W. B.
H. Lousberg and H. C. J. Ottenheijm, Tetrahedron, 1990, 46,
1745.
47 A. Arcadi, S. Cacchi and F. Marinelli, Tetrahedron Lett., 1989, 30,
2581.
30 F. X. Schmid, in Protein Structure: A Practical Approach, ed.
T. E. Creighton, IRL Press, Oxford, 1997, ch. 11, pp. 261–297.
31 (a) H. Tsunematsu, R. Isobe, K. Mizusaki, S. Makisumi and M.
Yamamoto, Org. Mass Spectrom., 1994, 29, 260; (b) H. Tsunemat-
48 A. I. Roshchin and N. A. Bumagin, Chem. Heterocycl. Compd.,
1999, 35, 171.
New J. Chem., 2002, 26, 834–843
843